Introduction to probability and data

hi this is debrupa nandi today is my second day with git

studies

observational - 1. collect data in a way that does not directly interfere with how the data arise(“observe”) 2. only establish an association 3. retrospective uses past data 4. prospective - data are collected throughout the study

experiment- randomly assign subjects to treatments

Confounding variables - extraneous variables that affect both the explanatory and the response variable and that make it seem like there is a relationship between them

Correlation does not imply causation Observational staements help make us make correlation statements Experiments help us infer causation

In statistics, a confounder is a variable that influences both the dependent variable and independent variable, causing a spurious association. Confounding is a causal concept, and as such, cannot be described in terms of correlations or associations.